KR20220083831A - Image creation method, device, and storage medium - Google Patents

Abstract

An image generating method and apparatus, a device, and a storage medium related to the technical field of a computer are provided. The method includes: obtaining contour information and target area information; determining at least one target contour according to the contour information and the target region information, wherein all or part of each target contour is located in the target region; reducing first coordinates of the plurality of contour points in each target contour to obtain second position information of each target contour, wherein the second position information of the target contour is the second position information of the plurality of contour points in the target contour 2 coordinates, the magnitude of the reduction of the coordinates of each contour point equal; and generating a target image corresponding to the target region according to the target region information and the second position information of each target contour, the target image including contours located in the target region. The method reduces the size of the target image and further ensures the accuracy of the generated target image.

Description

Image creation method, device, and storage medium

This application is based on Chinese Patent Application No. 202010097623.0, entitled "IMAGE GENERATION METHOD AND APPARATUS, DEVICE, AND STORAGE MEDIUM", filed on February 17, 2020 priority, the entire contents of which are incorporated herein by reference.

TECHNICAL FIELD The present disclosure relates to the field of computer technology, and more particularly to methods, devices, and storage media for generating images.

Currently, the processing of whole slide images (WSIs) is increasingly based on deep learning network models. For example, before recognizing a target object in WSI, a network model must be trained using WSI and target images of WSI, where the target image is an image generated according to position information of an outline in WSI. Since the size of the target image is relatively large and exceeds the processing range of the current image generation technology, it is necessary to reduce the size of the target image.

In the prior art, the position information of the contour is reduced in the WSI, and a target image of the WSI is generated according to the reduced position information. However, when there are two contours that are relatively close in WSI, after the location information is reduced, the two contours may overlap. As a result, in the generated target image, the two contours are connected to become one contour, thereby obtaining an erroneous target image.

Embodiments of the present disclosure provide an image generating method, device, and storage medium capable of improving the accuracy of a target image. Technical solutions include:

According to one aspect, a method of generating an image is provided, the method comprising:

obtaining contour information and target area information, wherein the contour information includes first position information of at least one original contour in the original image, and the first position information of any original contour is located in any original contour. first coordinates of a plurality of contour points in which the target area information indicates a target area in the original image;

determining at least one target contour according to the contour information and the target region information, any target contour being located wholly or in part in the target region;

reducing first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of any target contour is located in the arbitrary target contour second coordinates of the plurality of contour points, wherein the first coordinates of the plurality of contour points in the at least one target contour have the same reduction range; and

generating a target image corresponding to the target region according to the target region information and second position information of the at least one target contour, the target image including the contour located in the target region.

According to a further aspect, there is provided an image generating device, the image generating device comprising:

an information acquiring module, configured to acquire the contour information and the target area information, wherein the contour information includes first position information of at least one original contour in the original image, and the first position information of any original contour is first coordinates of a plurality of contour points in the contour of , wherein the target region information indicates a target region in the original image;

a determining module, configured to determine at least one target contour according to the contour information and the target region information, any target contour being located in whole or in part in the target region;

a reduction module, configured to reduce first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of the arbitrary target contour is second coordinates of the plurality of contour points in the contour, wherein the first coordinates of the plurality of contour points in the at least one target contour have the same reduction range; and

and a generating module, configured to generate a target image corresponding to the target region according to the target region information and the second position information of the at least one target contour, the target image including the contour located in the target region.

In a possible implementation, the first coordinates comprise horizontal coordinates and vertical coordinates, the apparatus comprising:

obtain a first minimum horizontal coordinate and a first minimum vertical coordinate according to the first position information of the at least one target contour; determine a first minimum horizontal coordinate as a decreasing range of horizontal coordinates of a plurality of contour points in the at least one target contour; and an extent obtaining module further configured to determine a first minimum vertical coordinate as a reduced range of vertical coordinates of a plurality of contour points in the at least one target contour.

In a possible implementation example, the first coordinates include horizontal coordinates and vertical coordinates, and the range obtaining module is configured to: obtain a second minimum horizontal coordinate and a second minimum vertical coordinate of the target area according to the target area information; determine a second minimum horizontal coordinate as a decreasing range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determine a second minimum vertical coordinate as a decreasing range of vertical coordinates of the plurality of contour points in the at least one target contour.

In a possible implementation embodiment, the determining module is further configured to, in response to determining that the arbitrary contour point is located in the target region according to the contour information and the target region information, determine an original contour to which the arbitrary contour point belongs as the target contour is composed of

In a possible implementation, the determining module comprises:

a reduction unit, configured to decrease the first position information of the at least one original contour to obtain third position information of the at least one first contour, any first contour after the corresponding original contour is reduced contour, wherein the first positional information of at least one original contour has the same reduction rate;

a first generating unit, configured to generate a first image corresponding to the original image according to the third position information of the at least one first contour, the first image including the at least one first contour;

a reduction unit, configured to reduce the target area information to obtain the first area information, wherein the first area information indicates a reduced target area, and a reduction rate of the target area information is a reduction rate of the first position information of the at least one original contour Same as -; and

and a first determining unit, configured to determine the at least one target contour according to the first area information and the first image.

In a possible implementation, the contour information further comprises a contour identifier of the at least one original contour; The first generating unit is configured to generate a first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour, the first image being the at least one first contour and the at least one the contour identifier of the first contour of , wherein the contour identifier of the arbitrary first contour is the contour identifier of the original contour corresponding to the arbitrary first contour.

In a possible implementation example, the first determining unit determines the contour identifier of the at least one second contour according to the at least one first contour, the at least one contour identifier of the first contour, and the first area information in the first image, - the at least one second contour is located wholly or in part in the reduced target area in the first image; and determine, as the target contour, an original contour corresponding to the at least one second contour according to the contour identifier of the at least one second contour.

In a possible implementation example, the first generating unit is configured to: set a pixel value in the contour area corresponding to the arbitrary first contour to the contour identifier of the arbitrary first contour; and generate the first image according to the third position information of the at least one first contour and the pixel value in the contour region corresponding to the at least one first contour.

In a possible implementation, the generating module comprises:

a second generating unit, configured to generate a third image according to the second position information of the at least one target contour, the third image including the at least one target contour; and

and a cutting unit, configured to cut the target area in the third image according to the target area information to obtain a target image corresponding to the target area.

In a possible implementation, the generating module comprises:

a second determining unit, configured to determine contour points located in the target area in the at least one target contour according to the target area information and second coordinates of the plurality of contour points in the at least one target contour,

The second determining unit is further configured to generate the target image according to the target area information and second coordinates of contour points located in the target area in the at least one target contour.

According to another aspect, there is provided a computer device, the computer device comprising a processor and a memory, the memory storing at least one instruction, wherein the at least one instruction is an operation performed in the method for generating an image according to the aspect described above. are loaded and executed by the processor to implement them.

According to another aspect, there is provided a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium storing at least one instruction, the at least one instruction generating an image according to the aspect described above. loaded and executed by a processor to implement the operations performed in the method.

According to another aspect, there is provided a computer program product, the computer program product storing at least one instruction, the at least one instruction being loaded by a processor to implement operations performed in an image generating method according to the above-described aspect. and executed.

The technical solutions provided in the embodiments of the present disclosure include at least the following beneficial effects.

According to the image generating method, device, and storage medium provided in the embodiments of the present disclosure, a target area is defined, so that the computer device only needs to generate a target image corresponding to the target area, and corresponding to other areas There is no need to create images. According to the present disclosure, the coordinates of a plurality of contour points with respect to the target contour are reduced to achieve the effect of cutting images corresponding to different regions, thereby reducing the size of the target image. Further, in the present disclosure, the coordinates of the plurality of contour points with respect to the target contour have a reduced range, ensuring that the distance between contours of the target image is equal to the distance between the corresponding contours in the original image, thereby , it avoids the problem of contour adhesion and ensures the accuracy of the generated target image.

Further, the minimum horizontal coordinate in the at least one target contour may be used as a decreasing range of the horizontal coordinate of each contour point in the at least one target contour, wherein the minimum vertical coordinate in the at least one target contour is in the at least one target contour can be used as the decreasing range of the vertical coordinates of each contour point in the You can avoid it as much as possible, and reduce the size of the target image.

Also, the minimum horizontal coordinate in the target area information may be used as a decreasing range of horizontal coordinates of each contour point in the at least one target contour, and the minimum vertical coordinate in the target area information is each contour in the at least one target contour It can be used as a reduction range of the vertical coordinate of the point, which is equivalent to using the minimum horizontal coordinate and the minimum vertical coordinate in the target area information as the origin of the contour coordinate system, avoiding drawing of the area outside the target area as much as possible, , it is possible to reduce the size of the target image.

Further, according to the present disclosure, the first positional information of the at least one original contour is further reduced to generate a first image corresponding to the original image. The first image includes a contour identifier of at least one original contour. The first image may be regarded as an index image of the original image. At least one target contour located in the target region in the original image may be determined according to the target region information of the original image and the first image, which increases the speed of determining the at least one target contour located in the target region, It also improves the efficiency of generating the target image.

In order to more clearly describe the technical solutions in the embodiments of the present disclosure, the following briefly introduces the accompanying drawings necessary for describing the embodiments. Of course, the accompanying drawings in the following description show only some embodiments of the present application, and a person skilled in the art may still derive other drawings from the accompanying drawings without creative efforts.
1 is a schematic diagram of an implementation environment according to an embodiment of the present disclosure;
2 is a flowchart of an image generating method according to an embodiment of the present disclosure;
3 is a schematic diagram of an original image according to an embodiment of the present disclosure;
4 is a schematic diagram of a target image according to an embodiment of the present disclosure;
5 is a schematic diagram of a target area of an original image according to an embodiment of the present disclosure;
6 is a schematic diagram of at least one target contour in a target area according to an embodiment of the present disclosure;
7 is a schematic diagram of a first image according to an embodiment of the present disclosure;
8 is a schematic diagram of a partial region of a first image according to an embodiment of the present disclosure;
9 is a schematic diagram of an outline of a first region according to an embodiment of the present disclosure;
10 is a schematic diagram of a partial region of another first image according to an embodiment of the present disclosure;
11 is a schematic diagram of a second image according to an embodiment of the present disclosure;
12 is a schematic diagram of another target image according to an embodiment of the present disclosure;
13 is a flowchart for generating a WSI target image according to an embodiment of the present disclosure;
14 is a schematic diagram of a comparison between a target image generated using conventional techniques and a target image using a method provided in the disclosure according to an embodiment of the present disclosure;
15 is a schematic structural diagram of an image generating apparatus according to an embodiment of the present disclosure;
16 is a schematic structural diagram of another image generating apparatus according to an embodiment of the present disclosure.
17 is a structural block diagram of a terminal according to an embodiment of the present disclosure;
18 is a schematic structural diagram of a server according to an embodiment of the present disclosure;

In order to make the objects, technical solutions, and advantages of the present disclosure more clear, implementation examples of the present disclosure will be described in more detail with reference to the accompanying drawings.

It may be understood that terms such as “first”, “second”, etc. used in this disclosure may be used to describe various concepts herein. However, unless otherwise specified, the concepts are not limited by the terms. Terms are only used to distinguish one concept from another. For example, first coordinates may be referred to as second coordinates, and similarly, second coordinates may be referred to as first coordinates, without departing from the scope of the present disclosure.

As used herein, the term “at least one” includes one or more, and the quantity of at least one is an integer. For example, at least one is 1, 2, 3, etc. As used herein, the term “plurality” includes two or more, and the quantity of the plurality is an integer. For example, the plural is 2, 3, 4, etc. As used herein, the term “each” refers to a plurality of corresponding each. For example, the plurality of contour points are three contour points, and each of the plurality of contour points refers to each of the three contour points. The term “optional” in this disclosure refers to any one of a plurality of corresponding ones. For example, the plurality of contour points are three contour points, and any one of the plurality of contour points refers to any one of the three contour points. For example, every contour point is a first contour point, or a second contour point or a third contour point.

The image generating method provided in the embodiments of the present disclosure is applicable to a computer device. In possible implementations, the computer device is a terminal such as a mobile phone, computer, or tablet computer. The terminal acquires the contour information and the target region information, and determines at least one target contour according to the contour information and the target region information, wherein any target contour is located in whole or in part in the target region; reduce the first coordinates of the plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour; Then, a target image corresponding to the target region is generated according to the second position information and the target region information of the at least one target contour, wherein the target image includes the contour located in the target region.

In another possible implementation, the computer device is a server. A server is one server, a server cluster including a plurality of servers, or a cloud computing service center. 1 is a schematic diagram of an implementation environment according to an embodiment of the present disclosure; Referring to FIG. 1 , the implementation environment includes a terminal 101 and a server 102 . The terminal 101 and the server 102 establish a communication connection through a wired or wireless network.

The target application may be installed in the terminal 101, and the target application may be an application having an image processing function or an application related to image processing. Server 102 is a server that provides services for target applications.

In a possible implementation example, the server 102 stores the contour information, the user sends the target area information to the server 102 through the target application of the terminal 101 , and the server 102 obtains the target area information . The server 102 determines at least one target contour according to the contour information and the target region information, wherein any target contour is located in whole or in part in the target region; reduce first coordinates of the plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour; Then, a target image corresponding to the target region is generated according to the second position information and the target region information of the at least one target contour, wherein the target image includes the contour located in the target region.

The method provided in this embodiment of the present disclosure is applicable to an image generation scenario. For example, in a scenario that creates a target image for WSI:

The WSI includes the target object, and the target image corresponding to the WSI includes the outline of the target object in the WSI. In some embodiments, the WSI and the target image corresponding to the WSI are used as training data of the network model, thereby training the network model capable of recognizing the target object in the WSI. By using the image generating method provided by the embodiments of the present disclosure, the coordinates of the contour points of each target contour may be reduced, thereby reducing the size of the generated target image. Further, the relative position of each contour point in the target image is not changed, i.e. the relative positional relationship between contours in the target image is the same as the relative positional relation between the corresponding contours in the original image, whereby, It avoids the problem of connecting the two contours of the image of , and guarantees the accuracy of the target image.

The image generating method provided in the embodiments of the present disclosure is applicable to all scenarios in which an image is generated, which is not limited to the embodiments of the present disclosure.

2 is a flowchart of an image generating method according to an embodiment of the present disclosure; The execution entity of embodiments of the present disclosure is a computer device. A computer device is a terminal, such as a mobile phone, computer, or tablet computer, or a server, or a server cluster comprising a plurality of servers, or a cloud computing service center. The image creation process is described in embodiments of the present disclosure. Referring to FIG. 2 , the image generating method includes the following steps.

201: Acquire contour information, wherein the contour information includes first position information of at least one original contour in the original image, and the first position information of any original contour includes a plurality of pieces in the original contour. first coordinates of the contour points of .

In an implementation, the original image is any image containing the target object. The original outline in the original image is the outline of the target object in the original image. As shown in FIG. 3 , the original image includes a plurality of target objects 301 , and the outlines of the target objects 301 are original outlines in the original image. In some embodiments, the target object is any object. For example, the original image is a pathological WSI, the target objects are cancer cells, the original contours are contours of cancer cells, or the target objects are a cell group of cancer cells, the original contour is a contour of a cell group and , a cell group is an inclusive portion of a plurality of cells.

In some embodiments, the at least one original contour refers to all original contours in the original image, or refers to some of the original contours in the original contour. This is not limited to the embodiments of the present disclosure.

As shown in FIG. 3 , the original image may include target objects, colors, textures, and the like. Fig. 4 shows a target image corresponding to the original image shown in Fig. 3; As shown in FIG. 4 , the target image contains no color or texture, only the original outline 401 of the plurality of target objects 301 . In an implementation, the original image and the target image corresponding to the original image are used as training data of the classification network model, and the classification network model may be used to distinguish target objects.

In some embodiments, the target image is generated according to first position information of at least one original contour in the original image. Accordingly, the target image corresponding to the original image has the same size as the original image. If the image size of the original image exceeds the first reference size, the size of the generated target image also exceeds the first reference size. For example, the first reference size is the maximum size of an image that can be created using current image processing techniques. Therefore, when the image size of the original image exceeds the first reference size, the target image corresponding to the original image cannot be generated using the current image processing technology. For example, the maximum size of an image that can be created using current image processing technology is determined by the processing upper limit of the programming language or the upper limit of the data that can be processed by the hardware device. The hardware device is a memory or the like.

For example, the size of an original image is 100,000*100,000, and the maximum size of an image that can be processed using current image processing techniques is 50,000*50,000. Therefore, the target image corresponding to the original image cannot be generated by the current image processing technology.

In addition, when the original image and the target image of the original image are used as training data of the network model, the original image and the target image of the original image must be smaller than the second reference size. When the sizes of the original image and the target image exceed the second reference size, the original image and the target image of the original image cannot be used as training data of the network model. For example, the second reference size is the maximum size of the training image used in the artificial intelligence algorithm. In an implementation, the second reference size is 2000*2000.

In the present disclosure, since the size of the generated target image can be reduced by defining a target area and generating a target image corresponding to the target area, the size of the original image is not limited in the present disclosure.

The first coordinates of the contour points are used to indicate the positions of the contour points in the original image. In an implementation, the first coordinates of the contour points are coordinates of the contour points in a two-dimensional coordinate system of the original image. In some embodiments, the two-dimensional coordinate system of the original image is a coordinate system in which the vertex of the original image is the origin of the coordinates. For example, the vertices of the original image are the vertices of the lower-left corner, the vertices of the lower-right corners, the vertices of the upper-left corners, the vertices of the upper-right corners, etc. of the original image, which are limited to embodiments of the present disclosure doesn't happen In an implementation, the two-dimensional coordinate system of the original image is the coordinate system used when the original image is generated, and any position point with respect to the original image is determined according to the two-dimensional coordinate system.

In a possible implementation, the contour information further comprises a contour identifier of the at least one original contour. The contour identifier is an identifier that uniquely determines the original contour. For example, the contour identifier is a sequence number, identity, ID, etc. of the original contour, and the contour identifier is not limited in the embodiments of the present disclosure. In an implementation, the contour identifier of the original contour and the first position information of the original contour are stored correspondingly. For example, the first position information of each original contour is stored in a file, and the file name of the file includes the contour identifier of the original contour.

In an implementation, the contour identifier is added to the first position information of the original contour. For example, the first position information includes first coordinates of a plurality of contour points in the original contour, and the first coordinates of the plurality of contour points are [[x_i_1, y_i_1], [x_i_2, y_i_2], [x_i_3] , y_i_3], ..., [x_i_n, y_i_n]] where x is the horizontal coordinate of the first coordinate, y is the vertical coordinate of the first coordinate, i is the contour identifier, and n is the contour point in the original contour. , where n is an integer greater than or equal to 3. In an implementation, i is the sequence number of the original contour, i = 1 means the original contour is the first contour, i = 2 means the original contour is the second contour, and i = n is the original contour It means that the contour is the nth contour.

For example, the method for acquiring contour information includes: acquiring contour information of the original image by performing contour analysis processing on the original image; To obtain contour information from a local storage space.

In an implementation, the computer device obtains contour information of the original image by analyzing and processing the original image. In a possible implementation, the original image comprises at least one target object; Obtaining the contour information of the original image by analyzing and processing the original image by the computer device, by the computer device, according to the pixel value of each point of the original image, the pixel value of any point and the adjacent point determining a pixel difference between pixel values; using any point as the contour point for the original contour in response to the pixel difference being greater than the reference difference; and obtaining position information of a plurality of contour points with respect to at least one original contour to obtain contour information of the original image.

In another possible implementation example, the original image includes at least one original contour, and obtaining contour information of the original image by analyzing and processing the original image by the computer device includes, by the computer device, the original contour. obtaining first coordinates of a plurality of contour points in the at least one original contour in a two-dimensional coordinate system of the original image according to a position of the at least one original contour with respect to the image; and and obtaining contour information of the original image according to first coordinates of a plurality of contour points in .

In an implementation, the contour information may alternatively be obtained from a local storage space, which is not limited in embodiments of the present disclosure. In an implementation, the local storage space includes contour information of the at least one original image, and the first location information of the at least one original contour of the original image is stored in the same storage location.

For example, the contour information is obtained by manually drawing the original contour on the computer device by the user, or is obtained by performing contour analysis processing on the original image by the computer device, which in the embodiments of the present disclosure not limited

In a possible implementation, the contour information is obtained by manually drawing an original contour on a computer device by a user. The computer device displays the original image, and the user draws the original image. The computer device generates an original contour corresponding to the sliding trajectory for the original image according to the detected sliding trajectory, and acquires first coordinates of a plurality of contour points in the original contour to obtain contour information.

In some embodiments, the user's drawing on the original image is a trigger operation on the original image. For example, the trigger operation is a click operation, a long press operation, a drag operation, a double click operation, a sliding operation, and the like. For example, the trigger operation is triggered by the user through a mouse or by touching the display screen by the user. In an implementation, the trigger operation on the original image is one or a combination of a plurality of trigger operations, which is not limited in the embodiments of the present disclosure.

In a possible implementation example, the contour information is obtained by performing contour analysis processing on the original image by the computer device. In an implementation embodiment, the original image includes at least one target object, and obtaining contour information of the original image by analyzing and processing the original image by the computer device includes, by the computer device, the contour information of the original image. determining a difference between the pixel value of any point and the pixel value of the adjacent point according to the pixel value of the arbitrary point and the pixel value of the adjacent point; using any point as a contour point for the original contour in response to a difference greater than a reference difference to determine the original contour of the target object; And, according to the positions of the plurality of contour points in the original contour in the two-dimensional coordinate system of the original image, first coordinates of the plurality of contour points in the original contour are obtained, and the plurality of contours in the original contour is obtained. and acquiring contour information of the original image according to the first coordinates of the points.

In an implementation, the original image includes at least one contour line of the original contour, and obtaining contour information of the original image by analyzing and processing the original image by the computer device includes, by the computer device, the contour information of the original image. obtaining first coordinates of a plurality of contour points in the at least one original contour in a two-dimensional coordinate system of the original image according to the position of the contour line of the at least one original contour with respect to the image of ; and acquiring contour information of the original image according to first coordinates of a plurality of contour points in one original contour.

A method of obtaining the contour information is only described by taking examples in the embodiments of the present disclosure, and is not limited to the above-described various obtaining methods. The contour information may alternatively be obtained in other manners. A manner of obtaining the contour information is not limited in the embodiments of the present disclosure.

202: Acquire target area information, wherein the target area information indicates a target area in the original image.

The target area information is information used to determine the target area. In a possible implementation, the target area information comprises coordinates of a plurality of vertices of the target area. For example, the target area is a rectangular area, and the target area information is coordinates of four vertices of the target area. In another possible implementation, the target area information includes the coordinates of any vertex of the target area, the shape and size of the target area. For example, the target area information includes that the shape of the target area is a rectangle, the length of the target area is 500, the width is 400, and the coordinates of the vertices in the upper left corner of the target area are (18000, 25000). That is, the shape, size, and position of the target area are all determined through target area information.

In an embodiment, the shape of the target area is fixed. For example, the shape of the target region is a rectangle, a circle, or the like, and the target region information includes coordinates of arbitrary vertices of the target region and the size of the target region. That is, the shape of the target region is the default, and the position and size of the target region are determined through target region information.

In an implementation, the target area information is input by a user or obtained by analyzing and processing the original image by a computer device. In a possible implementation example, the computer device displays a region information acquisition page, obtains target region information input by a user through the region information acquisition page, and the region information acquisition page is used to acquire target region information. For example, the area information acquisition page includes at least one area information input option.

In a possible implementation embodiment, the computer device divides the original image into a plurality of region images, performs contour recognition processing on each region image, and in response to the region image comprising the original contour, the region of the region image Use the information as target area information. For example, a region image including the original contour means that the region image contains the complete original contour, or the region image contains a portion of the original contour.

In some embodiments, the manner in which the computer device performs contour recognition processing on the area image to determine whether the original contour is included in the area image is a pixel value at any point on the area image and a pixel value at an adjacent point. It is for determining whether the original contour is included in the region image according to the difference between the two, or whether the original contour is included in the region image according to whether there are contour lines in the region image. Of course, other schemes may alternatively be used to determine whether the original contour is included in the area image, which is not limited in embodiments of the present disclosure.

For example, as shown in FIG. 5 , the area 501 of the original image includes a plurality of original contours. The computer device uses the area 501 as the target area and obtains area information of the area 501 as the target area information. However, region 502 of the original image does not include the original outline, and the computer device does not use region 502 as a target region.

203: Determine at least one target contour according to the contour information and the target region information, wherein any target contour is located in whole or in part in the target region.

When any target contour is fully or partially located in the target area, it is meant that any target contour is entirely located in the target area or partially located in the target area. That is, the target region includes a closed region corresponding to a complete target contour, or an unclosed region corresponding to an incomplete target contour. For example, as shown in FIG. 6 , the target area is a rectangular area, and the target area includes a plurality of target contours. The first target contour 601 surrounds the closed area, and the first target contour 601 lies completely in the target area. The second target contour 602 also surrounds the closed area, and the second target contour 602 is also located entirely in the target area. The third target contour 603 does not surround the closed area, and the third target contour 603 is located partially in the target area.

In a possible implementation, determining the at least one target contour according to the contour information and the target area information comprises, in response to determining that any contour point is located in the target area according to the contour information and the target area information, any and determining an original contour to which the contour point belongs to the target contour.

In an implementation example, the position information of the contour point may include a contour identifier of an original contour to which the contour point belongs, and the computer device may obtain the contour identifier included in the position information of the contour point. According to the contour identifier included in the positional information of the contour point, an original contour to which the contour point belongs is determined, and the determined original contour is determined as a target contour. That is, by obtaining the contour identifier of the original contour, the original contour is determined as the target contour.

For example, the coordinates of the contour point located in the target area are [x_2_n, y_2_n], where 2 is the contour identifier of the original contour. The contour identifier of the original contour to which the contour point belongs may be obtained according to the coordinates of the contour point. Then, according to the contour identifier, first position information of the original contour whose contour identifier is 2 may be obtained from the contour information.

In the above-described manner, whenever the target contour in the target region is determined according to the target region information, the contour points of each original contour must be retrieved and compared. If the target contour in each target region is to be determined according to a plurality of target region information, contour points of each original contour must be iteratively searched and compared.

Accordingly, in embodiments of the present disclosure, at least one target contour in the reduced target area may be further determined in the first image by generating the first image. The first image is a target image corresponding to the reduced original image.

In another possible implementation example, determining the at least one target contour according to the contour information and the target area information comprises: a first position of the at least one original contour to obtain third position information of the at least one first contour reducing information, wherein any first contour is a contour after the corresponding original contour has been reduced, and the first position information of the at least one original contour has the same reduction rate; generating a first image corresponding to the original image according to the third position information of the at least one first contour, the first image including the at least one first contour; reducing target area information to obtain first area information, wherein the first area information indicates a reduced target area, and a reduction rate of the target area information is equal to a reduction rate of the first position information of the at least one original contour -; and determining at least one target contour according to the first region information and the first image.

In some embodiments, reducing the first position information of the at least one original contour is reducing the coordinates of the plurality of contour points for each original contour. When the coordinates of the contour points include horizontal coordinates and vertical coordinates, the horizontal and vertical coordinates of the contour point must be reduced, where the horizontal coordinates and the vertical coordinates have the same reduction rate. For example, the horizontal coordinate of each contour point is reduced 10 times, and the vertical coordinate of each contour point is decreased 10 times, to obtain third position information of the at least one first contour.

In an implementation example, the generating of the first image corresponding to the original image according to the third position information of the at least one first contour includes setting a two-dimensional coordinate system, and the third position information of the at least one first contour determining the position of each contour point in the two-dimensional coordinate system according to and generating the first image in the two-dimensional coordinate system according to the position of each contour point in the two-dimensional coordinate system. Since the coordinates of each contour point are reduced, the generated first image is also reduced, and the reduction rate of the first image relative to the original image is equal to the reduction rate of the respective contour point coordinates.

In an implementation embodiment, the contour information further includes a contour identifier of the at least one original contour; The generating of the first image corresponding to the original image according to the third position information of the at least one first contour includes: the at least one third position information of the first contour and the contour identifier of the at least one original contour generating a first image according to the method, wherein the first image comprises at least one first contour and at least one contour identifier of the first contour, wherein the contour identifier of any first contour is any first contour is the contour identifier of the original contour corresponding to .

For example, as shown in FIG. 7 , in the first image, the region surrounded by the contour 701 includes the contour identifier 45 of the contour 701 , and the region surrounded by the contour 702 is the contour 702 . contains the contour identifier 47 of , the region surrounded by the contour 703 contains the identifier 48 of the contour 703 , and the region surrounded by the contour 704 contains the contour identifier 49 of the contour 704 . include

In an implementation, determining the at least one target contour according to the first region information and the first image comprises: in the first image at least one first contour, a contour identifier of the at least one first contour, and a first region determining a contour identifier of the at least one second contour according to the information, the at least one second contour being a first contour located wholly or partially in the reduced target area in the first image; and determining an original contour corresponding to the at least one second contour as the target contour according to the contour identifier of the at least one second contour.

For example, the target area information includes that coordinates of the vertices in the upper left corner of the target area are (30000, 36000), the length of the target area is 4000, and the width of the target area is 4000. After reducing the first position information of the original contour by 20 times, the computer device acquires third position information of the first contour, and generates a first image according to the third position information of the first contour. 8 shows a partial area of the first image. Since the first location information is reduced by 20 times, the target area information must also be reduced by 20 times to obtain the first area information. As shown in FIG. 8 , in the first region information, the coordinates of the vertices in the upper left corner of the first region 801 are (1500, 1800), the length of the first region 801 is 200, and the first region (801) includes a width of 200. According to the at least one first contour in the first image, the contour identifier of the at least one first contour, and the first region information, of the at least one first contour located wholly or in part in the reduced target region in the first image. It can be determined that the contour identifiers are 45, 47, and 49, respectively, ie the contour identifiers of the at least one second contour are 45, 47, and 49, respectively. 9 shows three contours in the first area 901 .

In an implementation, generating the first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour comprises: determining a pixel value in a contour region corresponding to any first contour setting a contour identifier of an arbitrary first contour; and generating the first image according to the third position information of the at least one first contour and the pixel value in the contour region corresponding to the at least one first contour. For example, the generated first image includes colors or includes three colors of black, gray, and white, which is not limited in embodiments of the present disclosure.

For example, FIG. 10 shows a partial area of the first image. 10 does not show different pixel values in the contour regions corresponding to different contours, but uses the letter i to denote the pixel value in the contour region, where i = 45 is the pixel value in the contour region 1001 45, i = 47 indicates a pixel value of 47 in the outline area 1002, i = 48 indicates a pixel value of 48 in the outline area 1003, and i = 49 indicates a pixel value of 47 in the outline area 1004. indicates that the pixel value is 49. By obtaining the pixel value of each contour region, the contour identifier of each target contour is obtained.

Via the first region information and the first image, at least one target contour located in the target region can be quickly determined, which improves the speed of determining the at least one target contour, so that the computer device can determine the first of the target contour. It is possible to draw the target image as quickly as possible according to the position information, thereby improving the efficiency of generating the target image.

In step 203 , the computer device obtains a contour identifier of the target contour. Prior to step 204 , the computer device may further obtain first position information of the at least one target contour from the contour information according to the contour identifier of the at least one target contour.

For example, the target area includes three target contours, and the contour identifiers of the three target contours are 45, 47, and 49, respectively. The contour information is retrieved according to 45 , 47 , and 49 to obtain first position information of the three target contours. The first position information of the three target contours is as follows.

[[x_45_1, y_45_1], [x_45_2, y_45_2], [x_45_3, y_45_3], ..., [x_45_n, y_45_n]],

[[x_47_1, y_47_1], [x_47_2, y_47_2], [x_47_3, y_47_3], ..., [x_47_n, y_47_n]],

[[x_49_1, y_49_1], [x_49_2, y_49_2], [x_49_3, y_49_3], ..., [x_49_n, y_49_n]].

204: reducing first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of the arbitrary target contour is and second coordinates of the plurality of contour points in the contour, wherein the first coordinates of the plurality of contour points in the at least one target contour have the same reduction range.

When the image corresponding to the target contour is generated according to the coordinates of the target contour, the size of the image corresponding to the target contour is a*b, where a is the maximum horizontal coordinate of the target contour, and b is the maximum of the target contour. is the vertical coordinate. Thus, the size of the image corresponding to the target contour is related not only to the size of the target contour, but also to the coordinates of the target contour. The coordinates of each contour point in the target contour are reduced to minimize unnecessary areas and to reduce the size of the generated image, thereby obtaining when generating an image corresponding to the target contour according to the coordinates of the target contour. Prevents the image size from becoming relatively large. For example, when the coordinates are reduced, the reduction range is any value greater than zero.

In a possible implementation, the first coordinates comprise horizontal coordinates and vertical coordinates, and reduce the first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour. Prior to this, the method includes: obtaining a first minimum horizontal coordinate and a first minimum vertical coordinate according to first position information of at least one target contour; determining a first minimum horizontal coordinate as a decreasing range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determining the first minimum vertical coordinate as a decreasing range of vertical coordinates of the plurality of contour points in the at least one target contour.

For example, in the first position information of the at least one target contour, the minimum horizontal coordinate (ie, the first minimum horizontal coordinate) is 5543, and the minimum vertical coordinate (ie, the first minimum vertical coordinate) is 4796 . (5543, 4796) is used as the reduction range, ie, (5543, 4796) is subtracted from the first coordinate of each contour point with respect to the target contour. If the first coordinates of the contour point are (6543, 5796), (5543, 4796) is subtracted from (6543, 5796), and the obtained coordinates are updated with the second coordinates of the contour point.

For example, the first position information, the first minimum horizontal coordinate, and the first minimum vertical coordinate of the at least one target contour satisfy the following relationship.

x_min = MIN[x_i_n];

y_min = MIN[y_i_n];

where x_min is a first minimum horizontal coordinate, y_min is a first minimum vertical coordinate, i is a contour identifier, n is a quantity of contour points of an arbitrary target contour, and n is an integer greater than or equal to 3; MIN[] is a function taking a minimum value, x is a horizontal coordinate, y is a horizontal coordinate, [x_i_n] is an arbitrary horizontal coordinate in the first position information of at least one target contour, and [y_i_n] is at least one It is an arbitrary vertical coordinate in the first position information of the target contour.

In some embodiments, reducing the first coordinates of the plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour comprises at least one according to a reduction range of the horizontal coordinates. reducing horizontal coordinates at first coordinates of a plurality of contour points in a target contour of ; and reducing vertical coordinates in the first coordinates of the plurality of contour points in the at least one target contour according to a reduction range of the vertical coordinates to obtain second coordinates of the plurality of contour points in the at least one target contour. includes steps.

In an implementation, the second position information of the at least one target contour, the first position information of the at least one target contour, the first minimum horizontal coordinate, and the first minimum vertical coordinate satisfy the following relationship.

contour_list_local = contour_list -[x_min, y_min]

where contour_list_local is a second coordinate list of contour points of at least one target contour, contour_list is a first coordinate list of contour points of at least one target contour; x_min is the first minimum horizontal coordinate, and y_min is the first minimum vertical coordinate.

That is, the origin of the coordinate system of the coordinates of the contours in the target area is set to [x_min, y_min], and the target image may be generated in the coordinate system based on contour_list_local.

A first minimum horizontal coordinate and a first minimum ordinate of the at least one target contour are used as a reduction range for reducing first coordinates of a plurality of contour points in each target contour, which reduce the image corresponding to the target contour. It is the same as cutting, thereby reducing the size of the target image corresponding to the target area.

In another possible implementation example, the first coordinates include horizontal coordinates and vertical coordinates, and first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour. Before reducing, the method includes: obtaining second minimum horizontal coordinates and second minimum vertical coordinates of a target area according to target area information; determining a second minimum horizontal coordinate as a reduced range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determining a second minimum vertical coordinate as a decreasing range of vertical coordinates of a plurality of contour points in the at least one target contour. That is, the horizontal coordinates in the first coordinates of the plurality of contour points in the target area must be subtracted from the second minimum horizontal coordinate of the target area, and the vertical coordinates in the first coordinates of the plurality of contour points in the target area are determined by the target area. The contour must be subtracted from the second minimum vertical coordinate of the target area, which is equivalent to using the vertex of the target area as the coordinate origin of the contour coordinate system and generating the target image as the coordinate origin, thereby reducing the size of the target image can do it

The second minimum horizontal coordinate and the first minimum horizontal coordinate may be the same or different, which relates to the actual situation of the first position information of the at least one target contour and the target area information. This is not limited in the embodiments of the present disclosure. Further, the second minimum vertical coordinate and the first minimum vertical coordinate may be the same or different, which relates to the actual situation of the first position information of the at least one target contour and the target area information.

205: Generating a target image corresponding to the target region according to the target region information and the second position information of the at least one target contour, wherein the target image includes a contour located in the target region.

In a possible implementation, generating the target image corresponding to the target region according to the second position information and the target region information of the at least one target contour comprises: generating a third image according to the second position information of the at least one target contour generating, wherein the third image includes at least one target contour; and cutting the target region in the third image according to the target region information to obtain a target image corresponding to the target region.

In some embodiments, cutting the target area in the third image according to the target area information to obtain a target image corresponding to the target area includes: coordinates included in the target area information to obtain updated target area information reducing them; determining a target area in the third image according to the updated target area information; and cutting the target area in the third image to obtain a target image corresponding to the target area. The reduction range of the coordinates included in the target area information is the same as the reduction range of the first coordinates of the plurality of outline points in the at least one target outline.

For example, when the target area information includes coordinates of a plurality of vertices of the target area, decreasing the coordinates included in the target area information reduces the coordinates of the plurality of vertices of the target area included in the target area information. refers to The reduction range of the coordinates of the plurality of vertices of the target area is equal to the reduction range of the first coordinates of the plurality of contour points in the at least one target outline.

For example, when the target area information includes the coordinates of an arbitrary vertex of the target area and the size of the target area, decreasing the coordinates included in the target area information may include coordinates of any vertex included in the target area information. are referred to as reducing In this case, the size of the target area included in the target area information remains unchanged.

In any case, after the coordinates included in the target area information are reduced, updated target area information may be obtained, and the updated target area information may indicate the target area in the third image. In other words, the target area in the third image may be directly determined according to the updated target area information, and then, the target image may be obtained by cutting the target area.

For example, a third image is shown in FIG. 11 . 11 includes complete contours of three target contours. 11 shows contour images of three target contours: a contour 1101 with a contour identifier of 45, a contour 1102 with a contour identifier of 47, and a contour 1103 with a contour identifier of 49. As shown in FIG. 11 further includes a target area 1104 . In FIG. 11 , the target area 1104 is cut to obtain a target image 1201 as shown in FIG. 12 .

In a possible implementation embodiment, generating a target image corresponding to the target area according to the target area information and the second position information of the at least one target contour comprises: second coordinates of a plurality of contour points in the at least one target contour and determining, according to the target area information, contour points located in the target area in the at least one target contour; and generating the target image according to the target area information and second coordinates of contour points located in the target area in the at least one target contour.

In some embodiments, determining the contour points located in the target area in the at least one target contour according to the target area information and second coordinates of the plurality of contour points in the at least one target contour comprises: reducing the coordinates included in the target area information to obtain the target area information; and determining contour points located in the target region in the at least one target contour according to the updated target region information and second coordinates of the plurality of contour points in the at least one target contour. The reduction range of the coordinates included in the target area information is the same as the reduction range of the first coordinates of the plurality of outline points in the at least one target outline.

In some embodiments, generating the target image according to the target area information and second coordinates of contour points located in the target area in the at least one target contour comprises: target area information to obtain updated target area information reducing the coordinates included in ; and generating the target image according to the second coordinates of the contour points located in the target region in the at least one target contour and the updated target region information. The reduction range of the coordinates included in the target area information is the same as the reduction range of the first coordinates of the plurality of outline points in the at least one target outline.

According to the image generating method provided in the embodiments of the present disclosure, a target area is defined, so that the computer device only needs to generate a target image corresponding to the target area, and to generate images corresponding to other areas. no need. According to the present disclosure, the coordinates of the plurality of contour points with respect to the target contour are reduced to achieve the effect of cutting images corresponding to different regions, thereby reducing the size of the target image. In addition, since the coordinates of a plurality of contour points with respect to the target contour in the present disclosure have the same reduction range, it is ensured that the distance between contours in the target image is equal to the distance between corresponding contours in the original image, thereby Thus, the problem of contour adhesion is prevented and the accuracy of the generated target image is ensured.

Further, the minimum horizontal coordinate in the at least one target contour may be used as a decreasing range of the horizontal coordinate of each contour point in the at least one target contour, wherein the minimum vertical coordinate in the at least one target contour is each in the at least one target contour can be used as a decreasing range of the vertical coordinates of the contour point of Much can be avoided, and the size of the target image can be reduced.

In addition, the minimum horizontal coordinate in the target area information may be used as a decreasing range of horizontal coordinates of each contour point in the at least one target contour, and the minimum vertical coordinate in the target area information is the minimum vertical coordinate of each contour point in the at least one target contour. It can be used as a reduction range of vertical coordinates, which is equivalent to using the minimum horizontal coordinate and minimum vertical coordinate in the target area information as the origin of the contour coordinate system of the target image, and it is possible to avoid drawing of the area outside of the target outline as much as possible. and the size of the target image can be reduced.

Further, according to the present disclosure, the first positional information of the at least one original contour is further reduced to generate a first image corresponding to the original image. The first image may include a contour identifier of the at least one original contour. The first image may be regarded as an index image of the original image. At least one target contour located in the target region in the original image may be determined according to target region information of the original image and the first image, which increases the speed of determining the at least one target contour located in the target region, and , it also improves the efficiency of target image generation.

The image generation method provided in the embodiments of the present disclosure may be applied to the process of processing pathological WSI. Currently, classification processing of pathological WSIs is increasingly based on classification network models. In order to use the classification network model to classify pathological WSIs, first the classification network model is trained, and then the trained classification network model is used to classify pathological WSIs.

When the classification network model is trained, the sample pathological WSI and the target image corresponding to this sample pathological WSI must first be acquired. For example, the sample pathological WSI includes a target object or target region, and a target image corresponding to the sample pathological WSI includes a contour of the target object or target region. In this embodiment, a sample pathological WSI including a target object is taken as an example for explanation.

In some embodiments, the sample pathological WSI comprises at least one target object, and the target contour corresponding to the sample pathological WSI comprises a contour of the at least one target object. The sample pathological WSI and target images are used as training sample data and fed into the classification network model. The classification network model processes the WSI to determine the contour of the target object in the WSI, determines whether the classification result is correct according to the contour in the target image corresponding to the sample pathological WSI, and then, the output result of the classification model By adjusting the parameters of the classification network model according to the difference between and the position information of the contour in the target image, the classification result of the classification network model can be more accurate. After the training of the classification network model is completed, the target object in WSI may be recognized based on the classification network model.

The size of the sample pathological WSI is relatively large, and the target image is the same size as the sample pathological WSI. Therefore, at present, scaled sample pathological WSI and scaled target image are used as sample training data. If two contours are close to each other in the scaled image, the two contours are glued to form one contour. Errors in the target image used to train the classification network model reduce the accuracy of the classification network model.

The present disclosure provides a method for generating a target image of a WSI. The method is applied to a computer device. As shown in FIG. 13 , the method of generating a target image of WSI includes the following steps.

1301: Obtaining contour information, wherein the contour information includes first position information of an original contour of at least one target object in the WSI, and the first position information of any original contour is in any original contour. first coordinates of a plurality of contour points in

In some embodiments, the WSI is an image containing the target object. For example, the target object is a cancer cell, a diseased blood vessel, a non-cancerous cell, a diseased blood vessel, and the like. The target object is not limited in the embodiments of the present disclosure.

1302: Acquire target area information, wherein the target area information indicates a target area in WSI.

1303: Determine at least one target contour according to the contour information and the target region information, wherein any target contour is located in whole or in part in the target region.

In a possible implementation, determining the at least one target contour according to the contour information and the target area information comprises, in response to determining that any contour points are located in the target area according to the contour information and the target area information, any and determining an original contour to which the contour point belongs to the target contour.

In a possible implementation example, determining the at least one target contour according to the contour information and the target area information comprises: first position information of the at least one original contour to obtain third position information of the at least one first contour reducing , wherein any first contour is a contour after the corresponding original contour has been reduced, and the first position information of at least one original contour has the same reduction rate; generating a first image corresponding to the WSI according to third position information of the at least one first contour, the first image including the at least one first contour; reducing target area information to obtain first area information, wherein the first area information indicates a reduced target area and a reduction rate of the target area information is equal to a reduction rate of the first position information of the at least one original contour; ; and determining at least one target contour according to the first region information and the first image.

In an implementation example, the contour information further includes a contour identifier of the at least one original contour, and generating the first image corresponding to the WSI according to the third position information of the at least one first contour comprises: generating a first image according to the third positional information of the first contour and the contour identifier of the at least one original contour, the first image comprising the at least one first contour and the contour identifier of the at least one first contour , and the contour identifier of the first contour is the contour identifier of the original contour corresponding to the first contour.

In an implementation, determining the at least one target contour according to the first region information and the first image comprises: in the first image at least one first contour, a contour identifier of the at least one first contour, and a first region determining a contour identifier of the at least one second contour according to the information, wherein the at least one second contour is a first contour located wholly or in part in the reduced target area in the first image; and determining an original contour corresponding to the at least one second contour as the target contour according to the contour identifier of the at least one second contour.

In an implementation, generating the first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour comprises: determining a pixel value in a contour region corresponding to any first contour setting a contour identifier of an arbitrary first contour; and generating the first image according to the third position information of the at least one first contour and the pixel value in the contour region corresponding to the at least one first contour.

1304: Decrease first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of the arbitrary target contour is and second coordinates of the plurality of contour points in the contour, wherein the first coordinates of the plurality of contour points in the at least one target contour have the same reduction range.

In a possible implementation, the first coordinates comprise horizontal coordinates and vertical coordinates, and reduce the first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour. Prior to this, the method includes: obtaining a first minimum horizontal coordinate and a first minimum vertical coordinate according to first position information of at least one target contour; determining the first minimum horizontal coordinate as a reduction range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determining the first minimum vertical coordinate as a decreasing range of vertical coordinates of a plurality of contour points in the at least one target contour.

In a possible implementation, the first coordinates comprise horizontal coordinates and vertical coordinates, and reduce the first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour. Prior to this, the method includes the steps of: obtaining a second minimum horizontal coordinate and a second minimum vertical coordinate of a target area according to target area information; determining a second minimum horizontal coordinate as a reduction range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determining the second minimum vertical coordinate as a decreasing range of vertical coordinates of a plurality of contour points in the at least one target contour.

1305: Generating a target image corresponding to the target region according to the target region information and the second position information of the at least one target contour, wherein the target image includes a contour located in the target region.

In a possible implementation, generating the target image corresponding to the target region according to the second position information and the target region information of the at least one target contour comprises: generating a third image according to the second position information of the at least one target contour generating, wherein the third image includes at least one target contour; and cutting the target region in the third image according to the target region information to obtain a target image corresponding to the target region.

In a possible implementation embodiment, generating a target image corresponding to the target area according to the target area information and the second position information of the at least one target contour comprises: second coordinates of a plurality of contour points in the at least one target contour and determining contour points located in the target region in the at least one target contour according to the target region information; and generating the target image according to the target area information and second coordinates of contour points located in the target area in the at least one target contour.

The target image corresponding to the target region of the WSI is generated using the above-described method for reducing the size of the generated target image. Thereafter, the image corresponding to the target region may be cut in the WSI according to the target region, and the image and the target image are used as training data of the classification network model. Since the distance between contours in the target image is the same as the distance between corresponding contours in WSI, the accuracy of the target image is guaranteed, thereby improving the accuracy of training of the classification network model.

14 is a schematic diagram of a comparison between a target image generated using conventional techniques and a target image generated using a method provided in this disclosure. As shown in FIG. 14 , in the target image 1401 generated using the prior art, the contour lines are blurred, and the contour lines are glued. The computer device erroneously judges the glued contour lines as one contour. However, the contour lines of the target image 1402 generated using the method provided in the present disclosure are clear, and the boundaries between the contour lines are clear, which prevents misjudgment by the computer device. Also, after target images corresponding to the plurality of target regions are generated, the plurality of target images may be combined according to a positional relationship between the target images, and may be combined into a target image corresponding to the WSI.

15 is a schematic structural diagram of an image generating apparatus according to an embodiment of the present disclosure; Referring to Figure 15, the device,

the information obtaining module 1501, configured to obtain the contour information and the target area information, the contour information includes first position information of at least one original contour in the original image, and the first position information of any original contour is comprising first coordinates of a plurality of contour points in any original contour, wherein the target region information indicates a target region in the original image;

a determining module 1502, configured to determine at least one target contour according to the contour information and the target region information, any target contour being located in whole or in part in the target region;

a reduction module 1503, configured to reduce first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of any target contour includes: a plurality of contour points in an arbitrary target contour, wherein first coordinates of the plurality of contour points in at least one target contour have the same reduction range; and

and a generating module 1504, configured to generate a target image corresponding to the target region according to the target region information and second position information of the at least one target contour, the target image including the contour located in the target region.

According to the image generating apparatus provided in the embodiments of the present disclosure, a target area is defined, so that the computer device only needs to generate a target image corresponding to the target area, and it is necessary to generate images corresponding to other areas there is no According to the present disclosure, the coordinates of a plurality of contour points with respect to the target contour are reduced to achieve the effect of cutting images corresponding to different regions, thereby reducing the size of the target image. In addition, since the coordinates of a plurality of contour points with respect to the target contour in the present disclosure have the same reduction range, the distance between contours in the target image is equal to the distance between corresponding contours in the original image, whereby The problem of contour adhesion can be prevented, and the accuracy of the generated target image is ensured.

16 , in a possible implementation, the first coordinates include horizontal coordinates and vertical coordinates, the apparatus comprising:

obtain a first minimum horizontal coordinate and a first minimum vertical coordinate according to the first position information of the at least one target contour; determine the first minimum horizontal coordinate as a decreasing range of horizontal coordinates of a plurality of contour points in the at least one target contour; and a range obtaining module 1505, configured to determine the first minimum vertical coordinate as a decreasing range of vertical coordinates of a plurality of contour points in the at least one target contour.

In a possible implementation example, the first coordinates include horizontal coordinates and vertical coordinates, and the range obtaining module 1505 is configured to: obtain a second minimum horizontal coordinate and a second minimum vertical coordinate of the target area according to the target area information; determine a second minimum horizontal coordinate as a decreasing range of horizontal coordinates of a plurality of contour points in the at least one target contour; and determine the second minimum vertical coordinate as a decreasing range of vertical coordinates of the plurality of contour points in the at least one target contour.

In a possible implementation, the determining module 1502 is responsive to determining that any contour point is located in the target region according to the contour information and the target region information, determining the original contour to which the arbitrary contour point belongs as the target contour. further configured to do so.

In a possible implementation, as shown in FIG. 16 , the determining module 1502 includes:

a reducing unit 1512, configured to reduce the first position information of the at least one original contour to obtain third position information of the at least one first contour, any first contour being reduced to a corresponding original contour a contour after being changed, and the first position information of at least one original contour has the same reduction rate;

a first generating unit 1522, configured to generate a first image corresponding to the original image according to the third position information of the at least one first contour, the first image including the at least one first contour;

a reduction unit 1512, further configured to reduce the target area information to obtain the first area information, wherein the first area information indicates a reduced target area, and a reduction rate of the target area information is determined by the second of the at least one original contour. 1 Same as reduction rate of location information -; and

and a first determining unit 1532, configured to determine the at least one target contour according to the first area information and the first image.

In a possible implementation, the contour information further comprises a contour identifier of the at least one original contour; The first generating unit 1522 is configured to generate a first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour, the first image being the at least one first contour and a contour identifier of at least one first contour, wherein the contour identifier of any first contour is a contour identifier of an original contour corresponding to the arbitrary first contour.

In a possible implementation example, the first determining unit 1532 is configured to: the contour of the at least one second contour according to the at least one first contour, the contour identifier of the at least one first contour, and the first area information in the first image determine an identifier, wherein the at least one second contour is a first contour located in whole or in part in the reduced target area in the first image; and determine, as the target contour, an original contour corresponding to the at least one second contour according to the contour identifier of the at least one second contour.

In a possible implementation example, the first generating unit 1522 sets the pixel value in the contour area corresponding to the arbitrary first contour to the contour identifier of the arbitrary first contour; and generate the first image according to the third position information of the at least one first contour and the pixel value in the contour region corresponding to the at least one first contour.

In a possible implementation, as shown in FIG. 16 , the generating module 1504 includes:

a second generating unit 1514, configured to generate a third image according to the second position information of the at least one target contour, the third image including the at least one target contour; and

and a cutting unit 1524, configured to cut the target area in the third image according to the target area information to obtain a target image corresponding to the target area.

In a possible implementation, as shown in FIG. 16 , the generating module 1504 includes:

a second determining unit (1534), configured to determine contour points located in the target area in the at least one target contour according to the target area information and second coordinates of the plurality of contour points in the at least one target contour,

The second generating unit 1514 is further configured to generate the target image according to the target area information and second coordinates of contour points located in the target area in the at least one target contour.

17 is a schematic structural block diagram of a terminal according to an embodiment of the present disclosure; The terminal 1700 is configured to perform the steps performed by the terminal or smart device in the above-described embodiments. For example, the terminal 1700 is a portable mobile terminal. Terminal 1700 may also be referred to as user equipment, a portable terminal, a laptop terminal, a desktop terminal, and the like.

In general, the terminal 1700 includes a processor 1701 and a memory 1702 .

Processor 1701 includes one or more processing cores, eg, a 4-core processor or an 8-core processor. Memory 1702 includes one or more computer-readable storage media. For example, a computer-readable storage medium is non-transitory. In some embodiments, the non-transitory computer-readable storage medium of the memory 1702 is configured to store at least one instruction, wherein the at least one instruction comprises an image generating method provided in embodiments of the present disclosure. and configured to be executed by the processor 1701 to implement.

In some embodiments, terminal 1700 may include, in implementations, a peripherals interface 1703 and at least one peripheral. The processor 1701 , the memory 1702 , and the peripheral interface 1703 may be connected by a bus or signal cable. Each peripheral may be connected to the peripherals interface 1703 by a bus, signal cable, or circuit board. For example, the peripheral includes at least one of radio frequency (RF) circuitry 1704 , display screen 1705 , camera component 1706 , audio circuitry 1707 , positioning component 1708 , or power source 1709 . do.

The peripheral interface 1703 may be configured to connect at least one peripheral related to input/output (I/O) to the processor 1701 and the memory 1702 . The RF circuitry 1704 is configured to receive and transmit RF signals, also referred to as electromagnetic signals. The display screen 1705 is configured to display a user interface (UI). The camera component 1706 is configured to acquire an image or video. Audio circuitry 1707 includes a microphone and a speaker. The microphone acquires sound waves from users and surroundings, converts the sound waves into electrical signals and inputs these signals to the processor 1701 for processing, or sends these signals to an RF circuit ( 1704). The speaker is configured to convert electrical signals from the processor 1701 or RF circuitry 1704 into sound waves. The positioning component 1708 is configured to determine a current geographic location of the terminal 1700 to implement a navigation or location-based service (LBS). The power source 1709 is configured to power the assemblies in the terminal 1700 .

A person skilled in the art will know that the structure shown in FIG. 17 does not limit the configuration of the terminal 1700, and the terminal 1700 may include more or fewer components than those shown in the drawing, or some components. It will be appreciated that these may be combined, or arrangements of other components may be used.

18 is a schematic structural diagram of a server according to an embodiment of the present disclosure; Server 1800 may vary widely due to different configurations or performance, and may include one or more central processing units (CPU) 1801 and one or more memories 1802 . The memory 1802 stores at least one instruction, and the at least one instruction is loaded and executed by the processor 1801 to implement the image generating method provided in the above-described method embodiments. Of course, the server 1800 may further have a wired or wireless network interface, a keyboard, an input/output interface, and other components to facilitate input/output. The server 1800 may further include other components for implementing device functions. Details are not described here. The server 1800 may be configured to perform steps performed by the server in the above-described image generating method.

An embodiment of the present disclosure further provides a computer device including a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction performs operations performed in the image generating method according to the above-described embodiments. loaded and executed by the processor to perform.

An embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing at least one instruction, wherein the at least one instruction performs operations performed in the image generating method according to the above-described embodiments. to be loaded and executed by the processor.

An embodiment of the present disclosure further provides a computer program or computer program product, wherein the computer program or computer program product stores at least one instruction, wherein the at least one instruction is performed in the image generating method according to the above-described embodiments loaded and executed by the processor to perform the specified operations.

A person skilled in the art may understand that all or part of the steps of the above-described embodiments may be implemented by hardware, or a program instructing related hardware may be implemented. The program may be stored in a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium is a read-only memory (ROM), a magnetic disk, or an optical disk.

The foregoing descriptions are merely optional embodiments of the present disclosure, and are not intended to limit the present disclosure. Any modification, equivalent replacement, or improvement made within the spirit and principle of the disclosure shall fall within the protection scope of the disclosure.

Claims (13)

A method for generating an image applicable to a computer device, the method comprising:
obtaining contour information and target area information, wherein the contour information includes first position information of at least one original contour in an original image, and the first position information of any original contour is first coordinates of a plurality of contour points in the contour, wherein the target area information indicates a target area in an original image;
determining at least one target contour according to the contour information and the target region information, any target contour being located in whole or in part in the target region;
reducing first coordinates of a plurality of contour points in at least one target contour to obtain second position information of the at least one target contour, wherein the second position information of any target contour is the arbitrary target contour second coordinates of a plurality of contour points in the at least one target contour, wherein the first coordinates of the plurality of contour points in the at least one target contour have the same reduction range; and
generating a target image corresponding to a target region according to the target region information and second position information of the at least one target contour, the target image including a contour located in the target region;
An image creation method applicable to computer devices. According to claim 1,
the first coordinates include horizontal coordinates and vertical coordinates, before reducing first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour In the method,
obtaining first minimum horizontal coordinates and first minimum vertical coordinates according to first position information of the at least one target contour;
determining the first minimum horizontal coordinate as a decreasing range of horizontal coordinates of the plurality of contour points in the at least one target contour; and
determining the first minimum vertical coordinate as a decreasing range of vertical coordinates of the plurality of contour points in the at least one target contour;
An image creation method applicable to computer devices. According to claim 1,
the first coordinates include horizontal coordinates and vertical coordinates, before reducing first coordinates of a plurality of contour points in the at least one target contour to obtain second position information of the at least one target contour In the method,
obtaining a second minimum horizontal coordinate and a second minimum vertical coordinate of a target area according to the target area information;
determining the second minimum horizontal coordinate as a reduced range of horizontal coordinates of a plurality of contour points in the at least one target contour; and
determining the second minimum vertical coordinate as a decreasing range of vertical coordinates of a plurality of contour points in the at least one target contour;
An image creation method applicable to computer devices. According to claim 1,
The determining of at least one target contour according to the contour information and the target area information may include:
in response to determining that the arbitrary contour point is located in the target region according to the contour information and the target region information, determining an original contour to which the arbitrary contour point belongs as the target contour;
An image creation method applicable to computer devices. According to claim 1,
The determining of at least one target contour according to the contour information and the target area information may include:
reducing the first position information of the at least one original contour to obtain third position information of the at least one first contour, wherein any first contour is a contour after the corresponding original contour has been reduced , the first positional information of the at least one original contour has the same reduction rate;
generating a first image corresponding to an original image according to third position information of the at least one first contour, the first image including the at least one first contour;
reducing target area information to obtain first area information, wherein the first area information indicates a reduced target area, and the reduction rate of the target area information is equal to a reduction rate of the first position information of the at least one original contour Same -; and
determining at least one target contour according to the first region information and the first image;
An image creation method applicable to computer devices. 6. The method of claim 5,
the contour information further includes a contour identifier of at least one original contour; Generating the first image corresponding to the original image according to the third position information of the at least one first contour comprises:
generating a first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour, the first image comprising the at least one first contour and the at least one first contour a contour identifier of , wherein the contour identifier of any first contour is a contour identifier of an original contour corresponding to any first contour;
An image creation method applicable to computer devices. 7. The method of claim 6,
The determining of the at least one target contour according to the first area information and the first image may include:
determining a contour identifier of at least one first contour in the first image, a contour identifier of the at least one first contour, and a contour identifier of at least one second contour according to the first region information; 2 contours are located wholly or partially in the reduced target area in the first image;
determining an original contour corresponding to the at least one second contour as a target contour according to the contour identifier of the at least one second contour;
An image creation method applicable to computer devices. 7. The method of claim 6,
generating the first image according to the third position information of the at least one first contour and the contour identifier of the at least one original contour,
setting a pixel value in the contour area corresponding to the arbitrary first contour to the contour identifier of the arbitrary first contour; and
generating a first image according to third position information of the at least one first contour and a pixel value in a contour region corresponding to the at least one first contour;
An image creation method applicable to computer devices. According to claim 1,
generating a target image corresponding to the target area according to the second position information and the target area information of the at least one target contour;
generating a third image according to second position information of the at least one target contour, the third image including the at least one target contour; and
cutting a target area in a third image according to the target area information to obtain a target image corresponding to the target area;
An image creation method applicable to computer devices. According to claim 1,
generating a target image corresponding to the target area according to the second position information and the target area information of the at least one target contour;
determining contour points located in a target region in at least one target contour according to target region information and second coordinates of a plurality of contour points in the at least one target contour; and
generating a target image according to target area information and second coordinates of contour points located in a target area in the at least one target contour;
An image creation method applicable to computer devices. A computer device comprising:
The computer device includes a memory and one or more processors, wherein the memory stores at least one instruction, wherein the at least one instruction is performed in the image generating method according to any one of claims 1 to 10. loaded and executed by the processor to implement operations;
computer device. A non-transitory computer-readable storage medium comprising:
The non-transitory computer-readable storage medium stores at least one instruction, wherein the at least one instruction is configured to implement operations performed in the image generating method according to any one of claims 1 to 10. loaded and executed by the processor;
A non-transitory computer-readable storage medium. A computer program product comprising:
The computer program product stores at least one instruction, the at least one instruction being loaded and executed by a processor to implement the operations performed in the method for generating an image according to any one of claims 1 to 10. felled,
computer program products.

Images